Parameter space result for the doped case

In this case one solves the two site problem for all doping cases at various local interaction values. The most prevalent differnce is the behavior of the hybridization to the bath site.

• 
• 
• 
• 
• 
• 
• 

from __future__ import division, absolute_import, print_function
import matplotlib.pyplot as plt
import numpy as np
from dmft.twosite_dop import dmft_loop_dop


def doping_config(res, u, name):
    fig, axes = plt.subplots(3, sharex=True)
    axes[-1].set_xlabel('$<N>_{imp}$')
    fill = res[:, 0]
    axes[0].set_xlim([0, 1])
    e_c = [sim.e_c for sim in res[:, 1]]
    V = [sim.hyb_V() for sim in res[:, 1]]
    mu = [sim.mu for sim in res[:, 1]]
    for feat, ax, lab in zip([e_c, V, mu], axes, ['$\\epsilon_c$', 'V', '$\\mu$']):
        ax.plot(fill, feat, label=lab)
        ax.set_ylabel(lab)
    axes[0].set_title('Two site DMFT parameters at U={}'.format(u))
    fig.savefig(name + '_bathparam.png', format='png',
                transparent=False, bbox_inches='tight', pad_inches=0.05)


def plot_doping_param(axis='real', beta=1e3, u_int=[4.]):
    for u in u_int:
        out_file = axis + '_dop_b{}_U{}'.format(beta, u)
        try:
            res = np.load(out_file + '.npy')
        except IOError:
            res = dmft_loop_dop(u)
            np.save(out_file, res)

        doping_config(res, u, out_file)

plot_doping_param(u_int=[2., 4., 5.85, 6., 8., 10., 100.])

Total running time of the script: ( 0 minutes 5.257 seconds)